Edit model card

indobert-finetune-tydiqa-transfer-indoqa

This model is a fine-tuned version of indolem/indobert-base-uncased on the tydiqa dataset. It achieves the following results on the evaluation set:

  • Loss: 2.3210

Model description

More information needed

Intended uses & limitations

More information needed

Training and evaluation data

More information needed

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

  • learning_rate: 1e-05
  • train_batch_size: 24
  • eval_batch_size: 24
  • seed: 42
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: linear
  • num_epochs: 10

Training results

Training Loss Epoch Step Validation Loss
3.162 1.0 241 3.2961
1.9671 2.0 482 2.4509
1.457 3.0 723 2.3005
1.2349 4.0 964 2.2628
1.0681 5.0 1205 2.1436
0.9338 6.0 1446 2.1978
0.8523 7.0 1687 2.2780
0.7909 8.0 1928 2.2481
0.725 9.0 2169 2.2912
0.6977 10.0 2410 2.3210

Framework versions

  • Transformers 4.23.1
  • Pytorch 1.12.1+cu113
  • Datasets 2.6.1
  • Tokenizers 0.13.1
Downloads last month
5

Dataset used to train andreaschandra/indobert-finetune-tydiqa-transfer-indoqa